That is, unfortunately, how science works in almost any nontrivial case - you get a number of studies that are contradictory or weak with simpler hypotheses, pe

That is, unfortunately, how science works in almost any nontrivial case - you get a number of studies that are contradictory or weak with simpler hypotheses, people modify them to account for the dissonance, run more experiments, rinse, repeat. It doesn't help that our observability of biological systems, while better than it's ever been before, is still p. bad in situ.